I-Corps: Scalable Artificial Intelligence-Supported Flood Resilience Assessment

I-Corps项目更广泛的影响/商业潜力是开发一个可扩展的洪水恢复软件框架，在各种洪水情景和多个地理尺度下提供快速、准确和有效的洪水损害预测。拟议的网络基础设施平台提供的服务可以提供可扩展的、动态的、智能的、建筑物级别的抗洪能力评估。该技术通过提供用户自定义场景的实时灾害预测服务，大大减少了测量房屋最低楼层高程的工作量，极大地促进了社区级洪水灾害评估。成功部署所提议的技术的一个好处可能是帮助社区快速探索洪水风险的空间分布，并测试不同强度的洪水事件如何影响单个房屋和整个社区。这些知识有望进一步使政府官员、急救人员和资源分配者受益。这些知识还将有助于提高区域和国家一级的洪水意识。I-Corps项目的基础是开发人工智能（AI）支持的地理空间网络基础设施平台，用于洪水灾害预测。现有社区一级的洪水恢复和适应能力往往以一种不可扩展的方式进行调查，使调查工作流程具有社区特异性，对其他社区或大地理范围的可转移性较低。相比之下，该技术实现了准确、快速、低成本的洪水恢复能力评估：1)利用美国国家建筑足迹和交叉参考洪泛平原产品，获得细粒度的建筑级洪水暴露；2)利用街景图像，提出可扩展的最低楼层高程检索工作流程；3)开发结合建筑特征和模拟洪水强度的洪水破坏模拟范式。4)设计一个可扩展的洪水恢复力评估在线门户，具有交互式更新、洪水情景选择、位置查询和报告生成的能力。拟议的人工智能支持的网络基础设施和洪水灾害模拟框架有望革新和转变大规模洪水灾害评估和洪水态势感知通信。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is the development of a scalable, flood resilience software framework to provide fast, accurate, and valid flood damage prediction under various flooding scenarios and at multiple geographic scales. The services provided by the proposed cyberinfrastructure platform may provide scalable, dynamic, intelligent, building-level flood resilience assessment. The proposed technology significantly reduces the workload of measuring house-level lowest floor elevation and largely facilitates community-level flood damage assessment by providing services of on-the-fly damage predictions with user-defined scenarios. One benefit of the successful deployment of the proposed technology may be to help communities quickly explore the spatial distribution of flood risks and test how flood events with varying intensities affect individual houses as well as the whole community. Such knowledge is expected to further benefit government officials, first responders, and resource allocators. The knowledge will also help promote flood awareness at the regional and national levels.This I-Corps project is based on the development of an Artificial Intelligence (AI)-supported geospatial cyberinfrastructure platform for flood damage prediction. Existing community-level flood resilience and adaptation are often investigated in an unscalable manner, making the investigation workflow community-specific with low transferability to other communities or to large geographical scales. In comparison, the proposed technology achieves accurate, fast, and low-cost flood resilience assessment by 1) deriving fine-grained, building-level flood exposure using United States national building footprints and cross-referenced floodplain products, 2) proposing a scalable workflow of lowest floor elevation retrieval, taking advantage of street view images, 3) developing a flood damage simulation paradigm incorporating building characteristics and simulated flood intensity, and 4) designing an online portal for scalable flood resilience assessment, with the capability of interactive updates, flood scenario selection, location queries, and report generation. The proposed AI-supported cyberinfrastructure and flood damage simulation framework are expected to renovate and transform large-scale flood damage assessment and flood situational awareness communication.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.